Nicole Marwell (left) and Julia Koschinsky (right), University of Chicago
This episode features two guests from the University of Chicago—Dr. Julia Koschinsky, the Executive Director for the Center for Spatial Data Science, and Dr. Nicole Marwell, an Associate Professor in the School of Social Service Administration. They are leading a project, funded by the Public Health National Center for Innovations (PHNCI), which is analyzing data on geographic access to health and human services to help government officials address gaps and maximize the impact of existing resources. The project will offer a replicable framework and tool for analyzing and improving distributions of public funds for health and human services.
Additional Resources:
- Email spatial@uchicago.edu to request access to the code on github for their open‐source spatial analytic web‐based tool
- Learn more about their PHNCI-funded project
- Read a blog about their design thinking process
- Visit the Center for Spatial Data Science website and follow @GeoDaCenter
- Visit the School of Social Service Administration website and follow @UChicagoSSA
This podcast is also available on iTunes, Stitcher, and TuneIn.
Takeaways from the Interview
In the words of Dr. Julia Koschinsky and Dr. Nicole Marwell…
1. Ask how existing systems can be extended for a larger purpose
“Part of this is about extending the systems that are designed to collect data for one purpose, like financial management of contracts, to be able to address larger and more system-wide questions like, is the money going to where the need is? E-procurement originally wasn’t designed to answer that question, but it could be a tool that could be extended to include the programmatic side.”
– Dr. Julia Koschinsky
2. Before starting data analyses, understand the questions you want to answer
“The thing that really has to drive this is, what is your question? What is the thing you want to know that can change practice and improve the lives of people on the ground? Sometimes we get a little seduced by all the fancy whiz-bangs of data analytics, but if you’re doing analysis that doesn’t answer a question that’s important and actionable, then to us, we’d rather not see that analysis done.”
– Dr. Nicole Marwell
3. Organizational change is often necessary to improve data quality
“I didn’t expect for about 80% of the effort to go into cleaning up the data, but it makes sense because the data reflect the decision processes, and if those are fragmented, you end up with fragmented data exhaust. Then you need organizational change, which is much harder to come by than running spatial analytics.”
– Dr. Julia Koschinsky